Retardation of Volatile Organic Compounds in Ground Water .../67531/metadc675569/m2/1/high_re… · Retardation of Volatile Organic Compounds in Ground Water in Low Organic Carbon

UCRL-ID-120471

Retardation of Volatile Organic Compounds in Ground Water in Low Organic Carbon Sediments

Fredric Hoffman

April 1995

This is an informal report intended primarily for internal or limited external distribution. The opinionsand conclusionsstated are those of the author and may or may not be those of the Laboratory. Work performed under the auspices of the U.S. Department of Energy by the Lawrence Livermore National Laboratory under Contract W-7405-Eng-48.

DISCLAIMER

Tlus document was prepared as an account of work sponsored by an agency of the United States Government. Ne~ther the United States Government nor the University of Cahfornia nor any of thew employees, makes any warranty, express or unphed, or assumes any legal habfity or responsibihty for the accuracy, completeness, or usefulness of any information, apparatus, product, or process dlsciosed. or represents that its use would not infnnge pnvatelv or\ned nghts. Reference herem to any spechc commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily conshtute or unply its endorsement, recommendahon, or favonng by the Uruted States Government or the University of Califorma. The views and opiruons of authors expressed herem do not necessaniy state or reflect those of the uruted States Government or the University of Califorma, and shall not be used for advertmng or product endorsement purposes.

This report has been reproduced directly kom the best available copy.

Available to DOE and DOE contractors from the Office of Scientific and Technical Morma tion

P.O. Box 62, Oak Ridge, TN 37831 Prices available from (615) 576-8201, FTS 6268401

Available to the public from the National Technical Information Service

U.S. Department of Commerce 5285 Port Royal Rd.,

Springtield, VA 22161

DISCLAIMER

Portions of this document may be illegible in electronic image products. images are produced from the best available original document.

Interim Report

Retardation of Volatile Organic Compounds in Ground Water

in Low Organic Carbon Sediments

Fredric Hoffman

April 1995

Approved by

Dr. Daniel Ronen

Feinberg Graduate School Weizmann Institue of Science

Retardation of Volatile Organic Compounds in Ground Water

in Low Organic Carbon Sediments

Introduction

All analyses of contaminant transport in ground water are very sensitive to the retardation

factor, i.e., the relationship of the linear velocity of the ground water to the linear velocity of the

contaminant. Equation (1) is used for calculating the retardation factor (R) (After Freeze and

Cherry, 1979).

where: Vw = velocity of the ground water

vc = velocity of the contaminant

pb = bulk density of the porous medium

n = porosity of the porous medium

Q = distribution coefficient

The variability of the bulk density/porosity term, in permeable unconsolidated sediments, is

much less than that of Kd, making Kd the dominant variable (Curtis et al., 1986). Kd is often

considered to be a measure of the sorption of the contaminant onto the matrix material and in these

circumstances is defined as the concentration of the contaminant on the solid in units of

weighvweight divided by the concentration of the contaminant in the liquid in units of weight

/volume. For those sediments containing >O. 1% organic carbon, absorption of volatile organic

compounds (VOCs) into the matrix of the organic materials present is recognized as the dominant

retarding mechanism (Karickhoff, 1984; Schwarzenbach and Giger, 1985). In sediments

containing 4.1% organic carbon, however, the mechanisms of apparent retardation are less well

understood .

F. Hoffman, Interim Report March 1995 1

I postulate that adsorption onto the surfaces of the aquifer matrix materials is only one of

the processes that results in the apparent retardation of contaminants in ground water in

unconsolidated sediments. A partial list of other possible mechanisms includes: hydrodynamic

dispersion; abiotic and/or biotic degradation (Curtis et al., 1986; Mackay et al., 1986); matrix

diffusion; partitioning to organic carbon; and diffusion into and entrapmenthetention of the

contaminant in constricted or dead-end pores. The aim of this work is to study the relative

importance of the above mentioned processes in defining the I(d of VOCs in sediments having a

low organic carbon content.

Retardation has been observed in the field as well as in the laboratory and influences the

leading edge of a VOC contaminant plume as well as the trailing edge of a plume during attempted

ground water remediation. There is empirical evidence from a number of remediation sites

(U.S.EPA, 1989) that demonstrates that aquifer cleanup by pump and treat techniques takes long

time periods, and that aquifers that appear to have been cleaned often become re-contaminated

following the cessation of pumping (Travis and Doty, 1990). There has also been a series of

carefully controlled experiments in the relatively homogenous sands of the Borden Aquifer in

Canada (Mackay et al., 1986) that have shed some light on sorption and retardation rates and

processes.

In addition, there have been a number of hypotheses that have been advanced to explain the

observed behavior of these chemicals. These include; chemical heterogeneities within aquifers

(Ronen et al., 1987), diffusion of contaminants into and out of heterogeneous geologic materials

(Hoffman, 1993a; Wilson, 1995), sorption/desorption hysteresis (Roberts et al., 1986; Domenico

and Schwartz, 1990), density induced flows (Shincariol and Schwartz, 1990; Hoffman et al.,

1993; Istok and Humphrey, 1993), contaminants left in zones of stagnation and in dewatered areas

within cones of depression in phreatic aquifers (Hoffman, 1993b), and the orientation of the axis

of the contaminant plume with respect to the location of the extraction and monitor wells

(Hoffman, 1993a). One, some, or all of these phenomena may be significant to the rate of

contaminant removal during aquifer cleanup.


The purpose of this research is to gain an understanding of the most important mechanisms

that control the retardation of VOCs in the absence of organic carbon and how the geologic

characteristics of the sediments influence the lunetics of these mechanisms. At the pore scale I will

attempt to discriminate between adsorption onto aquifer matrix materials and diffusion into low

mobility pore spaces. At the geologic feature scale, I will attempt to determine the relative

importance of large scale aquitards as possible sinks, during plume passage, and later as sources of

VOC contamination during aquifer cleanup. I will demonstrate that retardation of VOCs in low

organic carbon unconsolidated sediments is not simply the result of sorption, as has been assumed

up to now, but is indeed the result of a complex interaction of the mechanisms described above. In

fact, it may be that sorption plays only a minor role under these conditions and that diffusion

controlled mechanisms are dominant.

Progress to date

Column Experiment

An initial column experiment has been completed using several new technologies. This

experiment has provided baseline conditions and parameters that can be achieved in a laboratory

column, such as porosity, hydraulic conductivity, and Kd, and will be utilized in ongoing

experiments. The experiment also served to help develop our laboratory techniques to minimize

volatile losses during the experiment. Such techniques as the use of gas-tight syringes in the

syringe pump and improved sample collection techniques will improve the rigor of additional

experiments.

The column was designed to perform VOC retardation tests on geological materials. Since

the sorption tests were to be evaluated based on the retardation of the contaminant through the

column, the column was designed to approach complete and uniform contact of the contaminated

water with the sediment and to minimize preferential flow through the sediments or along the

boundaries of the column. In addition, the sediment was to be compacted to approximate field

E Hoffman, Interim Report March 1995 3

conditions. Another design constraint of the column was to minimize contaminant sorption on the

column construction materials.

The sediment used in the column was designed to approximate a low organic carbon, fine-

to medium-grained sand of known grain-size distribution. To achieve this, sand from the Mektesh

Gadol, in the Northern Negev Desert of Israel, was acid washed, sieved to three fractions, and the

three fractions were recombined in equal parts by weight. The bulk density and porosity of the

sand were determined gravimetrically and the hydraulic conductivity was measured with a constant

head flow test. The final bulk density was 1.84 g/cc, the porosity was 0.27, and the hydraulic

conductivity was 2.1 X 10-2 cmlsec.

To ensure that the column met its design requirements, flow tests were evaluated using

magnetic resonance imaging (MRI) (Hoffman et al., 1993). Three iterations of packing and

compaction efforts followed by MRI evaluations were necessary before we were satisfied that the

column had minimal preferential flow paths and minimal air bubbles.

The MRI images revealed the heterogeneities within the column. These heterogeneities

were expected even though I was trying to create a homogeneous media. The processes that caused

the heterogeneities during the column packing are the same processes that cause heterogeneous

deposition of natural sediments. Small but observable irregularities in the flow could be seen in

sequential images as the tracer passed through the column, and were interpreted as resulting from

small changes in hydraulic conductivities. Vertical slice images through the column also clearly

showed the effect of the differences in density between the water and the tracer solution. By

measuring the velocity of the tracer through the column and applying the known velocity of the

water to equation 2, I was able to calculate the retardation and lQ for the tracer. When the tracer

had completely invaded the column, additional images were taken as the column was flushed with

fresh water. Following 1.5 pore volumes of fresh water flow, some sorbed tracer could be seen

remaining within the finer-grained areas of the column. While the behavior of this polar metallic

ion can not be compared to that of the non-polar hydrophobic VOCs, that are the subject of this


study, this exercise does present new possibilities for the use of MRI in studying pore scale

contaminant transport.

An image of a slice through the center of the column is shown in Figure 1. Two of the

more prominent features of the packing process are labeled A and B on the figure. These

experiments demonstrated the potential value of MRI in conducting ground water contaminant fate

and transport studies in packed columns. The column packing and bubble avoidance and removal

techniques that were developed with the aid of the MRI will be used in our future column

experiments.

Figure 1. This MRI image was taken of the water saturated column before the introduction of the tracer. Note the prominent tight-colored coarser-grained feature (A) at the entrance of the column, and the darlt-colored finer-grained feature (B) about 3 cm into the column. Six cm of the sand are shown in this image.

The VOC used in the retardation experiment was trichloroethylene (TCE) with NaCl in180

depleted water as conservative tracers. The breakthrough curves for the TCE and the tracers were

compared and retardation and distribution coefficients were determined from equation 1.



Two experiments were conducted. TCE analysis of the water exiting the column was

accomplished using an on-line fiber-optic based sensor, chloride was analyzed by high pressure

liquid chromatography (HPLC), and 180 by mass spectrometry. When equation ( 1) was applied to

the results, the first experiment yielded a TCE Kd of 0.05 mug and the second experiment, 0.1 1

mug.

In order to provide a point of comparison for these column experiment values, I applied the

Kd determination method proposed by Piwoni and Banerjee (1989). This method is based upon

the octanol / water partition coefficient (&w) for sediments containing less than 0.1% total organic

carbon. This method is shown in equation (2), and Piwoni and Banerjee (1989) claim that it should

be accurate within a factor of two or three.

log Q = 1.01 log &w - 3.46 (2)

where: KOw = Octanol /Water partition coefficient

Applying the values of for TCE taken from Schwarzenbach and Giger (1985) to

equation (2) provided an estimated Kd for TCE of 0.07 mug. This value falls nicely between the

two values from the column experiments.

The results of the first experiment are shown in Figure 2. The failure of the TCE

breakthrough curve to achieve higher concentrations during the time frame of this experiment is the

result of rate limited sorption, sequestration of TCE due to other retarding processes, and volatile

losses during the sample collection.

The TCE analyses were conducted on line using a new fiberoptic sensor based on the rate

of change of light absorption by a reagent which changes color in the presence of TCE (Angel

et al., 1989; Milanovich et al, 1991). Use of the fiber-optic based sensor provided us with quick,

real-time TCE analyses that allowed us to establish column baseline parameters to be applied to the

more rigorous column experiments now underway.

c/co

1 .o

0.8

0.6

0.4

00

0.2

0.0

Figure 2. TCE and chloride breakthrough curves from the first TCE sorption experiment. Application of equation 1 yields an R of 1.34 and a Q of 0.05 mug.

Distribution coefficients from field investigation data

In an effort to develop a method to determine Kds in the field and to learn more about the

natural distribution of QS in a plume of contaminated ground water, the tetrachloroethylene (PCE)

and TCE data from 240 boreholes at Lawrence Livermore National Laboratory (LLNL) were

examined. The LLNL sediments are all low (<0.1%) in organic carbon. Saturated sediment

samples collected and analyzed during drilling, and water samples collected and analyzed following

the construction of wells in the same boreholes, were used in the analysis. The saturated sediments


-

were collected by the method published by Hoffman and Dresen (1990). This method allows the

collection of saturated sediment samples without cross-contamination from elsewhere in the vertical

profile of the borehole. The water samples are taken from the screened interval of the well that is

ultimately constructed in each borehole. The water samples and the saturated sediment samples that

came from the same screened interval were compared. The resulting values along with estimates of

the amount of porewater displaced by drilling fluids during drilling and the physical characteristics

of the sediment sample were processed by the mass balance equation (3) for IQ.

Kd = [((Css * (Pb + n)) - (Cw * n * Rpw)) / Pb] / Cw (3)

where: Css = concentration in saturated sediment @g/kg)

Pb = bulk density (kg/L)

n = porosity

Cw = concentration in well water (pg/L)

Rpw = % remaining pore water &er displacement / 100

This method has two significant uncertainties. The first concerns the question of how

representative the well water is of the porewater in the saturated sample. While the porewater

comes from a 75 mm long core sample, the well water sample is taken from a two to three meter

long screened area of the aquifer. In addition, the core sample is taken during drilling while the

water sample is taken following well development, days or weeks later. The second uncertainty

concerns the amount of core sample porewater that is displaced by the drilling fluid during drilling.

While I have not been able to fully resolve these uncertainties, I have attempted to address them.

To help decrease the uncertainty in the representativeness of the well water to the pore

water, I analyzed the variability of the well water with time. Selecting a high and a low value, I

calculated two u s and reported the range. To help resolve the issue of porewater displacement,

we extracted porewater from a core sample by centrifugation and analyzed the inorganic chemistry


and compared it to that of the drilling fluid and to well water. For the sample analyzed, there was

approximately 80% porewater displacement by the drilling fluid. This was the value I used in all of

the IQ calculations.

In spite of these remaining uncertainties, when all of the approximately 150 Kds for both

TCE and PCE were analyzed, they appeared to be lognormally distributed, and the harmonic mean

of the Kds (PCE 0.1 mVg; TCE 0.05 mVgr) agreed favorably with the I Q determined in the

column experiment and with the u s in low organic carbon sediments found at the Borden site

(Curtis et al., 1986).. In addition, there was close agreement between the Kds of the different

sediments, as described by the field geologist (Table 1.)

TABLE 1. Kd Determinations for PCE and TCE (80% Pore water displacement)

Gravel Sand silt All Data

PCE (132 Kds) Average Range 0.4- 1.16 0.25-0.76 0.21-0.71 0.27-0.8 1 Average 0.68 0.39 0.4 0.44 Standard Deviation 0.9 0.48 0.61 0.62 Harmonic Mean Range 0.05-0.27 0.03-0.17 0.02-0.19 0.03-0.19 Harmonic Mean 0.12 0.09 0.1 0.1

TCE (124 u s )

Average 0.31 0.27 0.16 0.27 Standard Deviation 0.32 0.33 0.2 0.32

Harmonic Mean 0.06 0.05 0.06 0.05

Average Range 0.22-0.5 1 0.18-0.50 0.10-0.34 0.19-0.27

Harmonic Mean Range 0.06-0.17 0.03-0.14 0.02-0.16 0.03-0.15

The Kds for the VoGs in the Borden aquifer materials (Curtis et al., 1986), the column

experiment described above, and the us estimated from the field investigation data also described

above, all fall within the range predicted by the Piwoni and Banerjee (1989) method. This finding

appears to indicate that VOC us in low organic carbon permeable aquifer materials are dependent

on the chemical characteristics of the VOC and independent of the geologic grain size of the aquifer

materials. This is contrary to the commonly accepted theory that retardation of VOCs in ground


water in low organic carbon sediments is primarily the result of sorption, and that sorption is a

function of surface area.

Multi-Laver Samp ler Experimeni

A preliminary multi-layer sampler experiment (DMLS) was performed in well W-415, a

contaminant extraction well at LLNL. The well is screened through the entire contaminated vertical

interval (25 m) and spans four permeable zones and the intervening lower permeability zones. In

this experiment, the four permeable zones were simultaneously sampled with cells at 10 em

intervals. The intervening low permeability zones were not sampled. The results of this experiment

are shown in Figure 3. This well is located near the source of the contamination and as expected

the contaminant concentrations were higher in the upper zones than in the lower ones. An

interesting feature of the second and third profiles in Figure 3 is that the PCE concentrations are

highest in the upper part of each zone and steeply decline with depth. One possible interpretation of

this decline is that the coarse-grained zones are being continually cleaned by the pumping well and

that new contamination continues to diffuse into the zone from the lower permeability zone,

immediately above.

This application of the DMLS is the first time that the instrument has been deployed in a

wire-wrapped well screen that extends through multiple permeable and low permeability geologic

zones. The quality of the profiles in Figure 3 provide encouragement that performing experiments

with the DMLS will provide data regarding the transport of VOCs across geologic contacts.

Program for further investigation

This research will consist of laboratory packed column experiments to examine and

compare the retardation rates for the leading and trailing edges of a VOC contaminant plume;

laboratory diffusion experiments to determine the rate of VOC diffusion into different geologic

materials; comparison of distribution coefficients ( u s ) for VOCs in low organic carbon sediments


estimated with the use of field data with those in the literature; and the field measurement of the

diffusion of VOCs from contaminated aquifers into adjoining aquitards.

G ravel1 y

gravel

24

25

26

27

20 35T

37 =s CL,

44

45

46

Sandy

_ _ 0 10 20 30 40 50 60 70 80 90 100 110

Lithology PCE (PPW

Figure 3. A profile of PCE in Well W-415 at LLNL. Data points are limited to the top four permeable zones.

A column experiment is underway, conducted under more rigorous conditions than the first

experiment. The column itself is a glass and teflon chromatography column, the apparatus has been

carefully designed and constructed to minimize volatile losses during the experiment, and the

analyses of the VOCs will be performed by gas Chromatography. The column will be packed with

a fine-grained, low organic carbon sand and four VOCs of differing octanol/water partitioning

coefficients will be simultaneously passed through the column at a concentration of 300 ppb each.

The chemicals are trichloroethylene (TCE), tetrachloroethylene (PCE), chloroform (CHCl3), and

carbon tetrachloride (CC4). The breakthrough curves for each of the VoCs will be plotted along

E Hoffman, Interim Report March 1995 ll

with a conservative tracer and the retardation of the leading edge of each of the individual plumes

will be calculated. When the column reaches or approaches equilibrium, the VOC containing

influent will be replaced with fresh water, the experiment will continue, and the cleanup curves will

be plotted. The experiment will proceed until the VOC concentrations are well out on the

asymptotic part of the cleanup curve. The retardation of the VOCs on the trailing edge of the

respective plumes will be calculated and compared to the leading edge retardation in an attempt to

observe and quantify any retardationhelease hysteresis.

A diffusion apparatus is being designed in an attempt to measure the diffusion of VOCs

dissolved in water into sediments of differing geologic characteristics. Continuing searches of the

literature have not yet uncovered any experimental work in this area, for VOCs under saturated

conditions. This experiment will direct water, containing the same four VOCs used in the column

experiment above, over vials packed with sands, silts, and clays. Five different sediments in

quadruplicate vials will be included in each of eight modular trays plumbed in parallel. Water

containing the VOCs will continually flow over the tops of the vials. At various times during the

experiment, probably at one month intervals, a tray will be removed from the system and the vials

analyzed for VOC content. The results of the experiment will be compared to the existing

theoretical models of diffusion of VOCs in porous media which include the diffusivity of the

chemical and tortuosity of the sediment. The objective will be to evaluate the relative magnitude of

diffusion into sediments of varying grain size.

A twenty meter long DMLS is being assembled for deployment in well W-415 at LLNL.

The DMLS will have cells at 5 cm intervals through the entire screened area of the well shown in

Figure 3. The purpose of this experiment will be to measure the transport of TCE and PCE across

the geologic contacts between the permeable zones and their bounding aquitards. I expect to

measure the extent of contaminant transport across these contacts to evaluate the importance of the

aquitards as sinks for contaminants during the contamination stage and as sources during the

cleanup stage.


- - A

Present understanding and hypotheses

Retardation of VOCs in low organic carbon sediments is commonly thought to be primarily

the result of sorption, and the studies at the Borden aquifer were all based on this assumption

(Curtis et al., 1986). While sorption may be one of the mechanisms of the apparent retardation of

contaminant plumes, there are a number of others that may be significant. Some of these are (not in

priority order): hydrodynamic dispersion, abiotic and/or biotic degradation, matrix diffusion,

partitioning to organic carbon, and entrapmenthetention of the contaminant in constricted or dead-

end pores in permeable sediments or in the slower moving waters in less permeable materials.

Since Kd is the only significant variable in the retardation equation (Eq. l), all of these

mechanisms must be included in any complete definition of Q. Sorption itself is a process that is

not simply described for non-polar hydrophobic compounds such as the VOCs studied in this

research. The conditions under which sorption can be described as an equilibrium or a non-

equilibrium process, remain controversial, as does the role of sorption/desorption hysteresis. The

hysteresis described here refers to the rate of desorption being slower than the rate of sorption. The

expression of this phenomena may be seen in the shape of the cleanup curve. When cleanup begins

there is a rapid decrease in contaminant concentration followed by a flattening or tailing of the

curve as it becomes asymptotic. In addition, the variability of contaminant concentration in the

ground water confounds the confident prediction of the effect of any of these processes.

VOCs are non-polar, or only slightly polar, and must compete with polar water molecules

for sorption sites on the charged surfaces of the aquifer solids. This characteristic of VOCs

explains why their ms are so low and why the understanding of their sorption kinetics is so much

more complex nhan that of polar metallic ion contaminants. It also helps to explain why the relative

importance of diffusion of VOCs into less mobile pores in the aquifer and into finer-grained

sediments within or bounding the aquifer, may be of major significance. It is possible that I will

find that sorption is only a minor factor in the retardation of VOCs under these conditions and that

the primary retarding factors are diffusion controlled.


Wilson (1995) has described tailing of contaminant concentrations during cleanup as being

partially attributed to slow bleeding of contaminants from aquifer heterogeneities controlled by

diffusion. Farrell and Reinhard (1994) have also suggested that there is rapid removal of

contaminants from the macroscale porosity between sediment grains, and slow removal controlled

by diffusion from the micropores that occur within grains. However, to date, the scale of the

geologic features which may sequester contaminants and then slowly allow them to re-enter the

ground water causing the asymptotic portion of the cleanup curve, has yet to be established.

My working hypothesis is that the tailing phenomena is caused by small scale

heterogeneities and not by large scale lenses within aquifers or by bounding aquitards, especially

as the plume moves farther away from its source. Near the source, the contaminants would have

passed through, and remained in, aquitards above the aquifer, creating a longer term source of

contaminants to the aquifer. Farther away from the source, the contaminants could only enter the

aquitards by transport from the aquifers. I plan to demonstrate, in the field, through the use of the

DMLS, and in the laboratory, through the column and diffusion experiments, that diffusion

limitations prevent significant sequestration of contaminants in large-scale geologic features, except

in the immediate vicinity of the source of the contaminants.

Laboratory studies have shown that VOCs have higher &s in clays than in silts, and

higher u s in silts than in sands (Bishop et al., 1989). However, from the results of my column

experiment, my work on estimating Kds from field data, and from the data at the Borden site

(Curtis et al., 1986) it appears that there is no significant difference between the Qs for permeable

sediments, with low organic carbon, of varying grain size. A possible explanation for this

phenomena is that aquifer materials contain enough fine-grained sediments or enough of a

controlling mineral that the distribution of the VOCs are "buffered" such that the Q appears to be

not affected by the variable grain size of the sediment. Alternatively, it may be that the density of

mesopores or micropores, proposed by Farrell and Reinhard (1994) may be very similar in all

aquifer materials regardless of their grain size, and that the Kd determined from retardation is

primarily a function of diffusion controlled sequestration of contaminants in these smaller pores.


Finally, sorptioddesorption of VOCs on low organic carbon sediments has been variously

reported as reversible (Karickhoff, 1984; Mackay et al., 1986) and as hysteretic (Bahr, 1989;

Domenico and Schwartz, 1990; Mackay et al., 1986). I suggest that one possible explanation for

this dichotomy is that the sorptioddesorption Q mechanism is indeed an equilibrium process and

is completely and instantaneously reversible, while the diffusion controlled Kd mechanisms

respond differently at different concentrations. It is probable that both of these mechanisms are at

work. The ability to discriminate between them and to measure the kinetics of each, is critical to

understanding the transport of VOCs in ground water.

References

Angel, S . M., Ridley, M. N., Langry, K., Kulp, T. J., Myrick, M. L. 1989. New developments

and applications of fiber-optic sensors. In: Murray, R.W., Dessy, R.E., Heineman, W.R.,

Janata, J., Seitz, W.R., eds. American Chemical Society Symposium Series 403. Washington,

DC: American Chemical Society. 23:345.

Bahr, J.M. 1989. Analysis of nonequilibrium desorption of volatile organics during field test of

aquifer decontamination. J. Contam. Hydrol., 4: 205-222.

Bishop, D.J., J. P. Knezovich, and D.W. Rice, Jr. 1989. Sorption studies of VCKs related to

soiUGround water contamination at LLNL. Lawrence Livermore National Laboratory,

Livermore, CA. (UCID-21651).

Curtis, G.P., Roberts, P.V., and Reinhard, M. 1986. A natural gradient experiment on solute

transport in a sand aquifer 4. Sorption of organic solutes and its influence on mobility. Water

Resources Research, Vol. 22, No. 13, pp. 2059-2067.

Domenico, P.A., Schwartz, F.W. 1990. Physical and chemical hydrogeology. New York: John

Wiley & Sons.

F. Hoffman, Interim Report March 1995

Farrell, J and Reinhard, M. 1994. Desorption of halogenated organics from model solids,

sediments, and soil under unsaturated conditions. 1. Isotherms and 2. Kinetics. Environ. Sci.

Technol. (28), 53-72.

Freeze, R.A., Cherry, J. A. 1979. Groundwater. New Jersey: Prentice-Hall, Inc.

Hoffman, F. 1993a. A strategy for improving pump and treat ground water remediation. in

Avogadro, A. and Ragaini, R.C. (eds.) Technologies for Environmental Cleanup: Soil and

Groundwater. Kluwer Academic Publishers.

Hoffman, F. 1993b, Ground Water Remediation Using “Smart Pump and Treat”. Ground

Water, Vol. 31 (1). p. 98-106.

Hoffman, F., and M. D. Dresen. 1990. A method to evaluate the vertical distribution of VOCs in

ground water in a single borehole. Ground Water Monitoring Review. 10(2), 95- 100.

Hoffman, F., 2. Pearl and M. Magaritz. 1993. Evaluation of flow characteristics of a sand column

using magnetic resonance imaging. Submitted to the Journal of Contaminant Hydrology, July

1993. Currently under revision.

Istok, J.D., and Humphrey, M.D. 1993. Buoyancy-induced flow in groundwater at small relative

densities: Implications for dispersivity determinations using tracer tests. Unpublished

manuscript. Oregon State University, Corvallis, OR.

Karickhoff, S . W. 1984. Organic pollutant sorption in aquatic systems. J. Hydraul. Eng., 110(6),

707-734.

Mackay, D. M., D. L. Freyberg, and P. V. Roberts. 1986. A natural gradient experiment on solute

transport in a sand aquifer: 1. Approach and overview of plume movement. Water Resources

Research, 22( 13), 2017-2029.

Milanovich, F.P., Daley, P.F., Langry, K., Colston Jr., B.W., Brown, S.B., Angel, S.M. 1991.

A fiberoptic sensor for the continuous monitoring of chlorinated hydrocarbons. Proc. Second

International Conference for Field Screening Methods for Hazardous Wastes and Toxic

Chemicals. Feb. 12-14, 1991. Las Vegas, NV: EPA p 43-48.


U.S. Environmental Protection Agency. 1989. Evaluation of groundwater extraction remedies.

EPA/504/0289/054.

Wilson, J. L. 1995. Removal of aqueous phase dissolved contamination: Non-chemically

enhanced pump and treat. in "Subsurface Restoration", H. Ward, J. Cherry, and M. Scalf

Eds., in press. Lewis Publishers, Boca Raton, Florida.

Piwoni, M. D., Bannerjee, P. 1989. Sorption of volatile organic solvents from aqueous solution

onto subsurface solids. J. Contam. Hydrol., 4: 162- 179.

Roberts, P.V., Goltz, M.N., Mackay, D.M. 1986. A natural gradient experiment on solute

transport in a sand aquifer: 3. Retardation estimates and mass balances for organic solutes.

Water Resources Research. Vol. 22: 13 p. 2047-2058.

Ronen, D., M. Magaritz, H. Gvirtzman, and W. Gamer. 1987. Microscale chemical heterogeneity

in groundwater. J. of Hydrol. 92, p. 173-178.

Schincariol, R. A. and Schwartz, F.W. 1990. An experimental Investigation of variable density

flow and mixing in homogeneous and heterogeneous media. Water Resources Research, Vol.

26, NO. 10: 2317-2329.

Schwarzenbach, R. P., Giger, W. 1985. Behavior and fate of halogenated hydrocarbons in

ground water. In: Ward, C. H., Giger, W., McCarty, P. L., eds. Ground Water Quality, New

Y ork Wile y-Interscience.

Travis, C. C. and C. B. Doty. 1990. Can contaminated aquifers at Superfund sites be remediated?

Environ. Sci. Technol., v. 24, no. 10.


Documents

Retardation of Volatile Organic Compounds in Ground Water .../67531/metadc675569/m2/1/high_re… · Retardation of Volatile Organic Compounds in Ground Water in Low Organic Carbon